RESUMO
BACKGROUND: This study aimed to stablish cut-off of early diagnosis of diabetic polyneuropathy (PDN) based on neuropathy symptom score (NSS) and neuropathy disability score (NDS); to determine the behavior of NDD and NDS in patients with and without PDN; and to verify the association between clinical and demographic variables with both tests. METHODS: This retrospective cohort included 86 patients with diabetes. The NSS and NDS evaluations were collected in medical records in two moments: initial (entry into service) and final (after three years). Individuals were categorized in three groups: G1- PDN in both evaluations (N = 27); G2- PDN only in the final evaluation (N = 16); G3-individuals without PDN (N = 43). A ROC curve was performed to evaluate the sensitivity and specificity of NSS and NDS for PDN diagnosis. ANOVA was used to compare NSS and NDS between groups and evaluations, and multiple regression was performed to find predictors of PDN. RESULTS: The NSS and NDS showed excellent sensitivity and specificity (NDS ≥1.5 and NSS ≥6.5) for PDN diagnosis. There was a significant difference between groups in initial (p = 0.000) and final (p = 0.000) NDS and NSS evaluations. There was an association between peripheral arterial disease (PAD) and increase in NSS (p = 0.024) in G2; and association between loss of protective sensation (LOPS) and increase in NSS in G3 (p < 0.001). CONCLUSION: NSS and NDS tests showed excellent sensitivity and specificity for early PDN diagnosis. Behavior of both tests can differ patients with and without PDN. Furthermore, PAD and LOPS can be a predictor of PDN evolution.
